Project Summary More than 400,000 people currently receive maintenance hemodialysis in the United States. While this therapy is life-extending, clinical outcomes and quality of life are poor. Current strategies to remove fluid with hemodialysis are imprecise, and events both during hemodialysis (e.g., intra-dialytic hypotension) and after (e.g., prolonged recovery time) are common complications associated with significant morbidity and mortality. Unfortunately, these events are challenging to predict with standard methods for hemodynamic monitoring. All of this highlights a need for greater precision in monitoring and improvement in fluid removal on hemodialysis. Continuous hematocrit monitors have been developed to non-invasively monitor relative changes in intravascular volume during a hemodialysis session. Yet, adoption of this technology into decisions on fluid management has been mixed. This is largely due to a lack of data to show a consistent association with intra- dialytic hypotension or improved clinical outcomes. However, studies have largely focused on plasma volume, but hypovolemia is not the only cause of hypotension. In order to leverage the full potential of this technology, we plan to incorporate changes in plasma volume with the rate of fluid removal during hemodialysis to yield a semi-continuous measure of the rate of vascular refilling from the interstitial tissue - termed plasma refill rate - as a dynamic measure of patient resilience. Examining patterns of plasma refill rate offers a unique opportunity to evaluate changes in patient resilience throughout a hemodialysis treatment that has yet to be fully explored. Additionally, the clinical impact of a new technology requires understanding of its ease of integration into the workflow. As such, for this work we will use two complementary approaches. Using a large dataset, we will examine the relationship between plasma refill rate and intra-dialytic hypotension by (1) determining if low plasma refill rates are associated with the number of hemodialysis sessions with intra-dialytic hypotension and (2) to develop a model using time-updated variables to predict the time to development of intra-dialytic hypotension. Concurrently, we will collect primary data to (3) evaluate the association between plasma refill rate and recovery time and (4) explore provider perspectives on the usability of this technology. Through these parallel projects we will improve our understanding of subtle hemodynamic changes that will inform future studies and fuel work towards precision volume management in maintenance hemodialysis. Moreover, if we find a critical role of plasma refill rate in the prediction of intra-dialytic hypotension or recovery time, we can leverage the semi-continuous nature of this technology to alter the user interface and improve delivery of care. In conjunction with formal coursework in a master's program for clinical epidemiology, the proposed application for the NRSA fellowship award will provide Dr. Wang with intensive training in advanced biostatistics, expertise in statistical modeling, primary data collection, and facilitate the establishment of a research niche in dialysis technology and end-stage kidney disease.